[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Some thoughts on kdelibs et al.
From:       Guillaume Laurent <glaurent () telegraph-road ! org>
Date:       2001-09-25 21:17:07
[Download RAW message or body]

On Tuesday 25 September 2001 21:37, Marc Mutz wrote:

> > And how often do you need this kind of information ?
>
> That's not the point. The point is that it _adds_ information, while
> stubbornly repeating the virtual keyword in each derived class doesn't.

Uhu... makes sense. Then again this information is already obtainable by a 
mere comparison of the headers.

> Worse, you could mistakenly add it to a non-virtual function (not sure
> if the compiler would complain <testing> no, it doesn't).

I wonder what happens then, though.

> You should let the compiler do the housekeeping.

I agree with this principle in general, and it's true as well that in many 
cases you can "guess" whether a method is virtual or not. But still I'm not 
fully convinced :-).

> And one last argument: If Bjarne himself doesn't repeat the virtual
> keyword... ;-)

Yes, but on very trivial examples, with the base class right above the 
derived ones.

-- 
					Guillaume.
					http://www.telegraph-road.org
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<

[prev in list] [next in list] [prev in thread] [next in thread]